---
library_name: transformers
base_model: Qwen/Qwen3-8B
tags:
- ronpo
- mnpo
- qwen3
- preference-optimization
---

# qwen3-8b-inpo-avg-eta0p005-s42

Research checkpoint for the RONPO AAAI revision experiments.

- Method: INPO-avg eta=0.005
- Base model: `Qwen/Qwen3-8B` with non-thinking generation protocol
- Seed: 42
- Local source at upload time: `/NHNHOME/WORKSPACE/26msit001_A/BASE/aipr_lab_sjkim_eval/revision_qwen3_8b/full_iter1/train/inpo_avg_eta0p005_s42_nhn-b200-q3-seq-07101430-inpo`
- Uploaded at UTC: 2026-07-10T16:05:55Z
- Run status metadata: `not found`

Qwen3-8B INPO baseline on the averaged three-reward oracle.

Intended use: reproducibility and evaluation for the RONPO paper. This
checkpoint is not intended as a production assistant.